Download Hydroxyl Group of a Phosphorylated Ribose

166
J. Am. Chem. Soc. 2000, 122, 166-167
Determination of the pKa of the 2′-Hydroxyl Group
of a Phosphorylated Ribose: Implications for the
Mechanism of Hammerhead Ribozyme Catalysis
Paul D. Lyne†,‡ and Martin Karplus*,†
Department of Chemistry and Chemical Biology
HarVard UniVersity, 12 Oxford Street
Cambridge, Massachusetts 02138
Physical and Theoretical Chemistry Laboratory
Department of Chemistry, Oxford UniVersity
South Parks Road, Oxford OX1 3QZ, England
ReceiVed June 1, 1999
ReVised Manuscript ReceiVed September 8, 1999
Hammerhead ribozymes1 are small catalytic domains found in
some viral satellite RNAs and are involved in RNA selfcleavage.2-4 The reaction catalyzed by hammerhead ribozymes
is a phosphate ester hydrolysis that is thought to proceed via
activation of a specific 2′-OH group by proton abstraction. The
resultant 2′-alkoxide acts as a nucleophile, attacking the electropositive phosphorus in an intramolecular fashion and displacing
the 5′-oxygen of the cleaved nucleotide (Figure 1). Hammerhead
ribozymes are metalloenzymes5,6 with an absolute requirement
for divalent metal ions for catalysis of the cleavage reaction.7
Magnesium ions are the natural cofactors for the hammerhead
ribozyme, but other divalent cations can replace magnesium. They
include Mn2+, Cd2+, and Co2+, which actually catalyze the
reaction at higher rates than Mg2+ under the same conditions,
and Ca2+, which catalyzes the reaction at a slower rate; a recent
study reports that very high concentrations of Li+ ions can replace
the Mg2+ ions.8 Cleavage rates are correlated with the pKa of the
solvated metal ion, with the rate increasing as the pKa of the
solvated ion decreases. This has traditionally been interpreted9
as indicating that a metal-coordinated hydroxide acts as the general
base in the reaction. An alternative interpretation of the data10
suggests that the mechanism involves two metal ions but that a
metal hydroxide ion is not the base in the reaction. The exact
number of metal ions involved in the reaction and the nature of
their interaction with the hammerhead ribozyme remains an area
of discussion.11 A knowledge of the pKa values of ionizable groups
at the active site of the ribozyme may aid in indicating possible
mechanistic pathways and in inferring the nature of the chemically
active species.12,13
Since deprotonation of the 2′-hydroxyl group is involved in
generating the attacking nucleophile in the hammerhead ribozyme
reaction, it is important to know the pKa of this group. The
experimental determination of pKa values in complex systems,
particularly the direct measurement of titrating groups of catalyti†
Harvard University.
Oxford University.
(1) Birikh, K. R.; Heaton, P. A.; Eckstein, F. Eur. J. Biochem. 1997, 245,
1-16.
(2) Buzayan, J. M.; Gerlach, W. L.; Bruening, G. Nature 1986, 323, 349353.
(3) Hutchins, C. J.; Rathjen, P. D.; Forster, A. C.; Symons, R. H. Nucleic
Acids Res. 1986, 14, 3627-3640.
(4) Prody, G. A.; Bakos, J. T.; Buzayan, J. M.; Schneider, I. R.; Bruening,
G. Science 1986, 231, 1577-1580.
(5) Pyle, A. M. Science 1993, 261, 709-714.
(6) Wilcox, D. E. Chem. ReV. 1996, 96, 2435-2458.
(7) Dahm, S. C.; Uhlenbeck, O. C. Biochemistry 1991, 30, 9464-9469.
(8) Murray, J. B.; Seyhan, A. A.; Walter, N. G.; Burke, J. M.; Scott, W.
G. Chem. Biol. 1998, 5, 587-595.
(9) Dahm, S. C.; Derrick, W. B.; Uhlenbeck, O. C. Biochemistry 1993,
32, 13040-13045.
(10) Pontius, B. W.; Lott, W. B.; von Hippel, P. H. Proc. Natl. Acad. Sci.,
U.S.A. 1997, 94, 2290-2294.
(11) Zhou, D.-M.; Taira, K. Chem. ReV. 1998, 98, 991-1026.
(12) Jencks, W. P. Catalysis in Chemistry and Enzymology; Dover: New
York, 1987.
(13) Fersht, A. Enzyme Structure and Mechanism, 2nd ed.; Freeman: New
York, 1985.
‡
Figure 1. (A) Schematic of the proposed mechanism for the first step
of phosphate ester hydrolysis in the hammerhead ribozyme (divalent
metals not shown). (B) Model phosphorylated ribose used in this study.
cally important residues in enzymes, is not a straightforward task.
Kinetic assignments of pKa values can be complicated by uncertainties in interpreting the pH dependence of the measured parameters.14 Moreover, different values are often obtained with different
techniques.15 An estimate16,17 of the pKa of the 2′-OH group of a
nucleotide is based on a value of 12.5 for the nucleoside 2′-anhydro-1-R-D-xylofuranosyluracil. However, the presence of the phosphate group could modify the pKa value. Usher et al. have
reported18 a value of 13.9 for the pKa of the 2′-OH group in cistetrahydrofuran-4-ol 3-phenyl phosphate, while Guthrie estimated19 the pKa of the 2′-OH group of nucleotides to be approximately 16 based on thermodynamic arguments. Kinetic pKa values
of 13 (at 333 K) have also been reported for dinucleotides.20
Since the value for the pKa of the 2′-OH group of a nucleotide
is not well established, a theoretical estimate of the pKa of the
2′-OH at the active site of hammerhead ribozymes is presented
here based on quantum chemistry calculations of a phosphorylated
ribose (Figure 1). The pKa of a given acid HA in aqueous solution
is related to the standard Gibbs free energy change of the
ionization reaction by the thermodynamic relationship
∆Gaq0
1
)
×
pKa ) -log Ka )
2.303RT 2.303RT
(∆Gg0 + ∆Gsolv0(H+) + ∆Gsolv0(A-) - ∆Gsolv0(HA)) (1)
corresponding to the thermodynamic cycle in eq 1.21 In eq 1, ∆Gg0
and ∆Gaq0 are the standard free energies of ionization in the gas
and in solution, respectively, and the ∆Gsolv0 values are the free
energies of solvation. All gas-phase free energies were calculated
using the Gaussian 94 program.22 Solvation energies were calcu(14) Knowles, J. R. CRC Crit. ReV. Biochem. 1975, 3, 165.
(15) Marcus, Y. Ion SolVation; Wiley: New York, 1985.
(16) Birnbaum, G. I.; Giziewicz, J.; Huber, C. P.; Shugar, D. J. Am. Chem.
Soc. 1976, 98, 4640-4644.
(17) Izatt, R. M.; Hassen, L. D.; Rytting, J. H.; Christensen, J. J. J. Am.
Chem. Soc. 1965, 87, 2760-2761.
(18) Usher, D. A.; Richardson, D. I., Jr.; Oakenfull, D. G. J. Am. Chem.
Soc. 1970, 92, 4699-4712.
(19) Guthrie, J. P. J. Am. Chem. Soc. 1977, 99, 3991.
(20) Jarvinen, P.; Oivanen, M.; Lonnberg, J. J. Org. Chem. 1991, 56, 5396.
(21) Kollman, P. Chem. ReV. 1993, 93, 2395.
(22) Gaussian 94 (revision D.1) Gaussian Inc.: Pittsburgh, 1995.
10.1021/ja991820i CCC: $19.00 © 2000 American Chemical Society
Published on Web 12/18/1999
Communications to the Editor
J. Am. Chem. Soc., Vol. 122, No. 1, 2000 167
Table 1. Solvation Free Energy Changes Involved in the
Ionization of MeOH and H3PO4a
∆Gaq
experiment36,37
B3LYP/6-31+G**
a
MeOH
MeO-
H3PO4
H2PO4-
-5.08
-5.50
-95.00
-93.03
-26.0
-20.8
-76.00
-79.83
All values in kcal/mol.
Table 2. Electronic and Solvation Energies for the Ribose (RH)
and Phosphorylated Ribose (pRH) and Their Conjugate Bases (Rand pR-) Calculated at the B3LYP/6-31+G** Levela
RH
RpRH
pRa
EElec
ZPE
G°(g)
∆Gs
-382.9107
-382.3388
-989.4385
-988.7286
0.1259
0.1107
0.16718
0.15209
-382.8161
-382.2584
-989.3113
-988.6156
-0.0213
-0.1205
-0.1089
-0.3423
G°(s)
pKa
-382.8374 13.1
-382.3789
-989.4202 14.9
-988.9579
Units are hartrees.
lated using the Poisson-Boltzmann method as implemented in
the Jaguar program.23,24 The value of the solvation free energy
of the proton was set equal to -262.23 kcal/mol.
Since pKa calculations depend strongly on solvation free
energies, accurate values for them are required. To address this
issue, certain of the parameters used in the standard Jaguar model23
were optimized for systems related to the one of interest here. In
particular, it is known that anions often pose problems for the
determination of accurate solvation energies with continuum
methods.25,26 The values of solvation energies obtained by using
a continuum dielectric method depend on the van der Waals radii
employed in the calculation. Several studies have used adjusted
van der Waals radii to obtain agreement with experiment.26-28
To determine the optimal set of radii for the molecules being
studied here, in particular the radii for the 2′-alkoxide oxygen
and the pro-R and pro-S oxygens of the phosphate group, the
solvation free energies of methanol and H3PO4 and their conjugate
bases were determined and compared with experimental data. The
calculated thermodynamic values are given in Table 1. It was
found that values of 1.42 and 1.26 Å for the phosphate oxygens
and methoxide oxygen, respectively, gave good agreement with
the available experimental data; the default radii in Jaguar for
these atoms are somewhat larger (1.60 and 1.45 Å, respectively).
The adjusted radii were used for the solvation calculations on
the model nucleotides.
As a test of the method we have calculated the pKa of the 2′hydroxyl of ribose. The relevant gas phase and solvation energies
are presented in Table 2. The pKa of a ribose using the optimized
radius for the alkoxide oxygen of the conjugate base is calculated
to be 13.14, which is rather close to the experimental value of
12.5; the value of the pKa calculated with the default Jaguar radii
is 18.45, showing the importance of consistent van der Waals
radii for continuum electrostatic calculations.26-28 The various
model calculations suggest that the basis set, the form of the
nonlocal functional, and the optimized radii for the PoissonBoltzmann calculations are appropriate for determining the pKa
of the phosphorylated ribose. The gas-phase electronic energies
and corresponding thermochemical data for the phosphorylated
ribose and its conjugate base are shown in Table 2. The pKa
calculated for the phosphorylated ribose (14.9) is considerably
higher than the only available experimental pKa (12.5) of a
(23) Jaguar, 3.5 ed.; Schrodinger, Inc.: Portland, OR, 1998.
(24) Tannor, D. J.; Marten, B.; Murphy, R.; Friesner, R. A.; Sitkoff, D.;
Nicholls, A.; Rignalda, M.; Goddard, W. A., III; Honig, B. J. Am. Chem.
Soc. 1994, 116, 11875-11882.
(25) Honig, B.; Sharp, K.; Yang, A.-S. J. Phys. Chem. 1993, 97, 1101.
(26) Stefanovich, E.; Truong, T. Chem. Phys. Lett. 1995, 244, 65-74.
(27) Lim, C.; Bashford, D.; Karplus, M. J. Phys. Chem. 1991, 95, 56105620.
(28) Sitkoff, D.; Sharp, K. A.; Honig, B. J. Phys. Chem. 1994, 98, 19781988.
nucleoside (see above).16,17,29 This is not surprising since the
conjugate base of that nucleoside is expected to be stabilized by
a hydrogen bond between the adjacent 3′-OH group and the
deprotonated 2′-O-. The pKa of a nucleoside that has no such
stabilization of the conjugate base should have a higher pKa.
Additionally, the molecule studied here has a negatively charged
phosphate group in close proximity to the 2′-OH group which
should make it thermodynamically less favorable for the 2′-OH
group to ionize than for the same group in the unphosphorylated
molecule. Thus, the upward shift by nearly 2 pKa units is
reasonable.
The large value for the pKa of the 2′-OH group found here
(pKa ) 14.9) has implications for the mechanism of hammerhead
ribozyme catalysis and more specifically for the role played by
metal ions in the reaction. It has been established, based on atomic
substitution of the pro-R and pro-S oxygens by sulfurs (the thioeffect30) and by spectroscopic studies,31 that a metal ion coordinates directly with the pro-R oxygen of the phosphate at the active
site.31 Kinetic studies of hammerhead cleavage have found that
there is an inverse relationship between the pKa values of solvated
metal ions and the rate of the cleavage reaction.9 This has been
interpreted to indicate that a metal hydroxide ion acts as the
general base in the reaction since the concentration of metal
hydroxide ions increases as the pKa of the hydrated metal ion
decreases. However, increasing the acidity of the hydrated metal
ion lowers the basicity of the metal hydroxide. This is expected
to lower its ability to deprotonate the 2′-OH group at the active
site of the hammerhead. The pKa calculated here for a 2′-OH
group in a nucleotide model is several units higher than the pKa
of several metal ions (e.g., Pb2+(7.7), Zn2+(9.0), and Mg2+(11.4)32)
that readily catalyze the reaction. It therefore seems unlikely that
metal hydroxide ions act as the general base in the reaction.10
Since a metal ion is known to bind directly to the adjacent pro-R
oxygen of the phosphate, it is conceivable that it promotes the
reaction by stabilizing the transition state. It is also possible that
it could bridge the 2′-oxygen and pro-R oxygen, which is
consistent with the. two metal ion model proposed for a number
of phosphoryl transfer reactions10,33,34 If so, it is expected that
the hydroxide bond at 2′-OH would be polarized and that the
pKa of the group would be reduced. The magnitude of the
polarization of the bond would be directly related to the identity
of the metal ion.
Since other ribozymes5 and ribonucleases35 employ mechanisms
that involve deprotonation of a 2′-OH at the active site, the result
presented here has implications for phosphate hydrolysis reactions
in many biological systems.
Acknowledgment. This work was supported in part by the Human
Frontiers in Science Program (P.D.L.), the Wellcome Trust (P.D.L.), and
National Institute of Health (M.K.). Calculations were performed at the
Pittsburgh Supercomputing Center, the Goddard Space Flight Center
(NASA), and the Oxford University Supercomputer Centre.
Supporting Information Available: Details are given of the methods
used for the calculation of the results in the text and tables (PDF). This
material is available free of charge via the Internet at http://pubs.acs.org.
JA991820I
(29) Saenger, W. Principles of nucleic acid structure; Springer-Verlag:
New York, 1984.
(30) Scott, E. C.; Uhlenbeck, O. C. Nucleic Acids Res. 1999, 27, 479-484.
(31) Cunningham, L. A.; Li, J.; Lu, Y. J. Am. Chem. Soc. 1998, 120, 45184519.
(32) Pan, T.; Long, D. M.; Uhlenbeck, O. C. DiValent Metal Ions in RNA
Folding and Catalysis; Pan, T., Long, D. M., Uhlenbeck, O. C., Eds.; Cold
Spring Harbor Press: Plainview, NY, 1993; pp 271-302.
(33) Steitz, T. A.; Steitz, J. A. Proc. Natl. Acad. Sci. 1993, 90, 6498-6502.
(34) Sawata, S.; Komiyama, M.; Taira, K. J. Am. Chem. Soc. 1995, 117,
2357-2358.
(35) Fersht, A. Structure and Mechanism in Protein Science; W. H. Freeman
and Company: New York, 1999.
(36) Cabani, S.; Gianni, P.; Mollica, V.; Lepori, L. J. Solution Chem. 1981,
10, 563-595.
(37) George, P.; Witonsky, R. J.; Trachtman, M.; Wu, C.; Dorwart, W.;
Richman, L.; Richman, W.; Shurayh, F.; Lentz, B. Biochim. Biophys. Acta
1981, 223, 1-15.